We propose to test the hypothesis that the long-term risk of coronary heart disease can be accurately predicted by the pressure of coronary calcific deposits, detected by screening with cardiac fluoroscopy. We will also evaluate whether this is a more sensitive method of screening (without loss of specificity) than is exercise testing. The two noninvasive tests chosen could easily be applied alone or together in a large-scale screening program and be used to test different aspects of the disease process (physiology versus anatomy). The utility of exercise testing for screening has been limited by its relatively low sensitivity since it can only detect atherosclerotic plaque large enough to significantly impede coronary flow. Sophisticated fluoroscopic techniques are more sensitive for detecting atherosclerotic plaque and probably for predicting coronary events as well. A single screen prospective epidemiological study involving both test will be conducted on 825 American veterans with at least a 15% Framingham risk probability of coronary heart disease which will be defined by the epidemiologic endpoints of medical history will be conducted at baseline, and subjects will be followed for six years with annual visits and contacts with the primary physicians to ascertain occurrence of coronary heart disease. The sensitivity and specificity of exercise testing and fluoroscopy in relation to the incidence of coronary heart disease will be determined and compared. The predictive utility of each test, relation to the six-year incidence of coronary events, independent of the other test and baseline risk factors, will be assessed by logistic analysis.